Background New safe and effective TB vaccine strategies to replace infant BCG vaccination are needed urgently. We evaluated the safety, reactogenicity and immunogenicity of three doses of the live-attenuated Mycobacterium tuberculosis (Mtb) vaccine candidate, MTBVAC, in comparison to BCG in South African newborns.

Methods Healthy infants, HIV-unexposed, BCG-naïve newborns without history of close TB contact were randomly allocated into three sequential cohorts to receive a single intradermal dose of BCG (SSI, 2.5×10⁵ CFU) or MTBVAC (2.5×10⁴; 2.5×10⁵ CFU; or 2.5×106 CFU).

Results 228 pregnant women consented, and 99 newborns were enrolled. Seventy-eight infants across all 3 cohorts had local reactions, all rated mild, except one grade 2 erythema. Induration, swelling, and erythema was more common with increased MTBVAC dosage. Induration and swelling were more common in MTBVAC 2.5x106 than in BCG and reactogenicity in MTBVAC 2.5x105 was same as BCG. Twelve infants experienced 14 vaccine-unrelated SAEs including one death due to bronchopneumonia. Eight infants commenced TB treatment for unconfirmed pulmonary TB (BCG n=4 and MTBVAC 2.5x104 CFU n=4) and one for unconfirmed TB meningitis (BCG). MTBVAC was highly immunogenic at all 3 doses, inducing predominantly Th1-cytokine-expressing CD4 T-cells, which peaked at day 56 and waned thereafter. The 2.5×105 and 2.5×106 CFU MTBVAC doses were more immunogenic than BCG, inducing very similar response magnitudes and phenotypes. Vaccination with any MTBVAC dose resulted in QFT conversion in most infants at Day 56, but these responses waned and reverted to QFT-negative in more than half by the end of the 1-year follow-up period.

Conclusion MTBVAC appeared safe and well tolerated and immunogenic at doses between 2.5×10⁴ CFU and 2.5×106 CFU in South Africans newborns. The 2.5×105 CFU MTBVAC dose was selected for the ongoing phase 3 trial in a high TB prevalence setting.